Jiadong Zhu.
← All publications

TPAMI 2025

Rethinking Efficient and Effective Point-Based Networks for Event Camera Classification and Regression

Hongwei Ren, Yue Zhou*, Jiadong Zhu*, Xiaopeng Lin*, Haotian Fu, Yulong Huang, Yuetong Fang, Fei Ma, Hao Yu, Bojun Cheng

Asterisk: Equal contribution.

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 47(8), 6228–6241 (2025)

PDF BibTeX DOI
EventMamba architecture: event-cloud downsampling, hierarchical feature extraction, and classification or regression (Figure 3).
EventMamba architecture: event-cloud downsampling, hierarchical feature extraction, and classification or regression (Figure 3).

Overview

EventMamba processes sparse event streams as point clouds, combining hierarchical spatial feature extraction with temporal aggregation and Mamba-based sequence modeling. The framework supports action recognition, camera pose relocalization, and eye tracking while keeping computational requirements low.

Event CameraPoint CloudMambaClassification and Regression
BibTeX citation
@article{ren2025rethinking,
  title = {Rethinking Efficient and Effective Point-Based Networks for Event Camera Classification and Regression},
  author = {Ren, Hongwei and Zhou, Yue and Zhu, Jiadong and Lin, Xiaopeng and Fu, Haotian and Huang, Yulong and Fang, Yuetong and Ma, Fei and Yu, Hao and Cheng, Bojun},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2025},
  volume = {47},
  number = {8},
  pages = {6228--6241},
  publisher = {IEEE},
  doi = {10.1109/TPAMI.2025.3556561},
  url = {https://doi.org/10.1109/TPAMI.2025.3556561}
}
Download citation ↓